Welcome to our community
Be a part of something great, join today!
Smoke on MAC
- Thread starter Frank Cueto
- Start date
Oliver Peters
Well-known member
- Joined
- Apr 5, 2009
- Messages
- 263
- Reaction score
- 0
- Points
- 0
Here's another link.
http://digitalfilms.wordpress.com/2009/12/18/smoke-on-the-mac/
I added some of the review I wrote about Smoke on Linux earlier in the year, just for general info. At this point, Smoke on Mac has Action, but not Batch. That means the compositing environment is still quite healthy for an NLE. The hardware I/O is limited by the Kona3 card, so that would be 1920x1080 or maybe 2K. Internal limit is 4K. Apple doesn't support the NVIDIA SDI daughter card, therefore, no Smoke Deliverables module. Also no Sparks support as yet.
- Oliver
http://digitalfilms.wordpress.com/2009/12/18/smoke-on-the-mac/
I added some of the review I wrote about Smoke on Linux earlier in the year, just for general info. At this point, Smoke on Mac has Action, but not Batch. That means the compositing environment is still quite healthy for an NLE. The hardware I/O is limited by the Kona3 card, so that would be 1920x1080 or maybe 2K. Internal limit is 4K. Apple doesn't support the NVIDIA SDI daughter card, therefore, no Smoke Deliverables module. Also no Sparks support as yet.
- Oliver
I would say Batch and Sparks is the most crucial here. It's ok without realtime deliverables. But autodesk said it's not that they dont want to include it in. Like sparks, it takes time for the 3rd party to catch up (which i believe they will), I hope Batch is going to be integrated in the future (this i'm not so sure). It's actually important for finishing system like smoke to have batch, i guess the only way to work this out is render it out as a clip and put it back.
Emery Wells
Well-known member
- Joined
- Feb 28, 2007
- Messages
- 705
- Reaction score
- 0
- Points
- 0
I think you guys forget that SMOKE on Linux didn't have batch until 2009, and it was still a very capable product.
Here is why SMOKE makes sense for our facility right now.
1. We need a machine that can work with high quality DPX files seamlessly. For color management this is critical. Dealing with Quicktime is a color management nightmare.
2. We need a system where we can work with our assembled timeline and do basic compositing very quickly (graphics, time remaps, paint, tracking, and roto). It must also be suitable for client supervised sessions.
So far we've eliminated Final Cut, Avid, After Effects, Nuke, Shake, Premiere, Combustion, and pretty much everything else on the market that would fall in the 'desktop' category.
Smoke was/is primarily about assembling all the elements together from all the different departments/specialty apps and still being able to make tweaks/changes as well as add finishing touches that need to happen once everything is all together.
If you do not have this need, than Smoke is not for you. If I needed compositing, Id choose Nuke. If I needed to do motion graphics, Id choose After Effects. If I needed to Edit, Id choose Final Cut or Media Composer.
My theory is that Smoke Linux will either be phased out or dropped to a software only product like it is on the Mac. At this point, who is going to buy a Smoke Advanced system on Linux? From my perspective, Autodesk has revealed a very clear roadmap and that is to move away from the hardware. As of today, there is absolutely nothing special about Autodesks hardware. It's off the shelf parts with a HUGE markup. First they launched Spark but require a Flame dongle, then SMACK, and I'll bet Spark, Lustre, and even Flame will be not too far behind. This is a roadmap that will take 3-5 more years to complete. Does this completely change Autodeks business in the professional market? Yes, but if they want to stay relevant in the world of Nuke, Scratch, and After Effects, they will need to adapt to survive.
I think its very positive to see them making these moves before the absolute last minute. Avid is a great example. They moved too slowly and allowed Final Cut to gain significant market share. Today, they are in the final stages of their 'turnkey hardware' transformation and they have started to regain mindshare from the Final Cut generation.
Anyone who wants the best performance, features, flexibility (access to sparks, for instance..), connectivity, and storage choices. That includes a lot of high end shops, who if they were adding to their Smoke inventory would likely not choose to go to a Macintosh, particularly if they have a Linux based infrastructure. Like Microsoft, Apple does not always play well with others, so buying into the Mac world means buying into Apple's proprietary stuff as well as their more open stuff. Pure IT people don't usually want to do that when Linux offers a much more open alternative that can be very useful when trying to do more automated processes through scripting, or when trying to create storage architectures that offer the best performance.
I suppose that's one way to look at it. But another way is to say that by having a very tightly controlled hardware base, support - a major issue in high end facilities that need near 24/7 uptime - becomes a lot more manageable. The problem that all PC developers face is the myriad of hardware combinations, even when they issue lists of "approved" combinations. Consumers and some professional users will put systems together that they "think" should work without even looking at those lists, and then be very upset when they don't. One advantage of the Mac for Autodesk is that there are a finite number of hardware combinations compared to the more general PC world, so that's a plus. But it's still not the same as issuing hardware with the software that the company KNOWS will work, has tested, and can offer phone support for that's meaningful. Selling the hardware is more about support management at this point than it is about making money on the hardware sale (that was a bit less true when the products were on SGI hardware, which was unique at the time).
I don't see where Smoke competes against any of those products. Nuke is a great compositor, but it isn't client friendly in most cases. Scratch is a fine conforming box and color corrector, but it doesn't have anything like the visual effects capabilities of Autodesk products, and is, in my experience, never used for that (unless the operator is a masochist). After Effects is a desktop product that does many things, but is at its heart a motion graphics program that can also be used for general compositing. But I can't see anyone using it in a client oriented commercial finishing session.
If what you're saying is that Autodesk should move towards lower priced software only distribution and leave the hardware up to the user, that probably is something they're looking at. But I certainly wouldn't make the assumption that Apple is their only likely target for that.
MichaelHalsell
Well-known member
I wonder if Autodesk ever considered renaming or branding the new Smoke for Mac product prior to release. As it stands, it seems there is a divide with MacSmoke vs LinuxSmoke. The camps seems to divided on those that know what "the Real Smoke is and is capable of doing" vs "The Mac version-which might be considered Junior Varsity".
Emery Wells
Well-known member
- Joined
- Feb 28, 2007
- Messages
- 705
- Reaction score
- 0
- Points
- 0
The case I made at the beginning of my post was that none of those products are a replacement for Smoke/Flame. However, over the next couple of years it's likely we'll see a desktop product compete in the Autodesk space. From a programming perspective, Nuke isn't any less sophisticated, they just have built a program for a different purpose. Whereas Smoke/Flame are built for working as quickly as possible with integrated toolsets, Nuke is built for best in class film pipeline compositing where granular control is more important than speed.
The market isn't going to support their pricing structure for too many more years. If The Foundry decided to compete (and they are the only company i see that is really well positioned to do so), Autodesk would have a serious problem on their hands. Autodesk knows it which is why they are taking steps to evolve with the changing market.
Certainly. I would expect multi platform support (Mac, Windows, Linux). I don't think Macs are better at everything Mike... just most things
There is certainly a large user base that would have reason to go for the Smoke Advanced on Linux but I will promise you will be a diminishing number over the next two years. Any facility that is considering a new Smoke today (be it an established facility with significant Autodesk investment or otherwise) will have to seriously consider SMACK as an option. When you can put together 3 SMACK systems for the price of one Smoke Advanced system, you'll have to have some very good reasons to spend the extra $$. Let's not forget most Smoke and Flames often have a Mac as an assist station sitting right next to it that's used for Photoshop, After Effects, whatever. If you could drop the linux box and do it all on the Mac, that makes a very compelling case.
The Mac version is almost identical to the base Smoke product on Linux. There is nothing "junior varsity" about it, other than the inability to use Batch (not in the base Linux product, either) and Real Time Deliverables (a limitation of the platform at this point in time). Those are not deal breakers unless they're things you use every day, and most compositing on Smoke can be done without Batch. In fact, there are a number of "old school" Flame operators who don't use it either.
Emery Wells
Well-known member
- Joined
- Feb 28, 2007
- Messages
- 705
- Reaction score
- 0
- Points
- 0
One other point Mike. We recently demo'd a full Linux setup that had Smoke, Flame, and Lustre loaded on it. The administration required to run one of these boxes is insane. We had a Flame admin (who had been doing this for 10+ years) at our shop for 2 full days trying to get everything up and running.
The install procedure is insane. You need a dedicated Flame admin if you plan on running a couple suites. Technical administration is a part of running any facility, but I almost think Autodesk makes it as complicated as possible just to keep the 'allure' of a big, expensive, complex system. It somehow helps justify the price.
The install procedure is insane. You need a dedicated Flame admin if you plan on running a couple suites. Technical administration is a part of running any facility, but I almost think Autodesk makes it as complicated as possible just to keep the 'allure' of a big, expensive, complex system. It somehow helps justify the price.
Really? From where I'm sitting, they don't have an editing product and they don't have anything even approaching a finishing product. They'd either have to acquire something or do a pretty serious development effort to get something that approaches Smoke - or Flame, for that matter.
I think you're clinging to a long held belief that has been somewhat tested and proven to not be as true as people sometimes think. The advent of Final Cut didn't kill Avid, it largely created a new market that wasn't there before. The advent of Color didn't kill Baselight, Lustre, Resolve, and other high end players. And the advent of Red didn't kill Arri, Panavision, Panasonic, and Sony. The growth of new markets doesn't necessarily destroy existing players, even though it might make their mission a bit more challenging. It sometimes creates downward pricing pressures, but the need for high end tools with high end performance still remains in the markets that it existed in previously. You likely won't find a lot of "A" title pictures having DI's done on Apple Color, and you won't find high end commercials being finished on After Effects, regardless of the price. Adjusting a pricing structure doesn't always mean you lower the price to $2000. What it does mean is that you try to find a sweet spot in which the additional sales balance against the lower price per seat, to the point that you're ultimately making more than you did selling fewer seats at a higher price. I think Autodesk is basically floating that now, trying to see exactly what price point they need to arrive at to make Smoke as a product as profitable (or close to it) as it has been up to this point. I think they've picked a good starting point for that approach. It might fluctuate either up or down depending on how many additional sales it generates, and ultimately they will have their "sweet spot" answer. If it works, I would expect some other players to do the same thing. You can draw your own conclusions as to who some of those "other players" might be.
Certainly. I would expect multi platform support (Mac, Windows, Linux). I don't think Macs are better at everything Mike... just most things
I get the impression you're saying that as a user, not as a systems engineer...:smile5:
Some do. Maybe that's more prevalent on the East Coast. Personally, I think the more compelling reason to purchase a Mac based Smoke is to have access to ProRes codecs, which are commonly used by commercial editors, a lot of whom are on Final Cut these days.
MichaelHalsell
Well-known member
So would there ever be need or request to have MacSmoke have complete parity with it's Linux counterpart? Other than have Spark support, is the product acceptable enough as it stands?
Until more of them are in the field, tested under battlefield conditions, and proven to be as reliable under those conditions as the Linux version, it's all a bit speculative. But pending that, my opinion would be yes, it is more than acceptable.
Hi Emery
Sure smoke is still a very capable product without Batch, but that was before. Since smoke is supposed to enable us to do work faster and quicker so having Batch is a plus with the recent introduction. That is the idea of new product. I just feel it doesnt make sense right now because they need to keep the integrity on linux version for the time being...but it's also due to technical limitation.
I'm not sure how they are going to separate the linux and mac version. As it stands right now, it still makes sense for the price point due to the feature differences. But say after sparks and batch are included in the future, linux version will seem less attractive anymore.
With this being said, I wonder when SDI board is going to be enable in future OS X.
it is..no sdi monitoring possible without that.
Bal Deo
Well-known member
There are some great tips here smoke on mac, including conforming red one files on smoke.
http://area.autodesk.com/blogs/discreetuk
You will need to register, the information here is just priceless.
http://area.autodesk.com/blogs/discreetuk
You will need to register, the information here is just priceless.